Ng, R.W.M., Shah, K., Specia, L. orcid.org/0000-0002-5495-3128 et al. (1 more author) (2016) Groupwise learning for ASR k-best list reranking in spoken language translation. In: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 20-25 Mar 2016, Shanghai. http://dx.doi.org/10.1109/ICASSP.2016.7472853, 2016-M . , pp. 6120-6124. ISBN 9781479999880
Abstract
Quality estimation models are used to predict the quality of the output from a spoken language translation (SLT) system. When these scores are used to rerank a k-best list, the rank of the scores is more important than their absolute values. This paper proposes groupwise learning to model this rank. Groupwise features were constructed by grouping pairs, triplets or M-plets among the ASR k-best outputs of the same sentence. Regression and classification models were learnt and a score combination strategy was used to predict the rank among the k-best list. Regression models with pairwise features give a bigger gain over other model and feature constructions. Groupwise learning is robust to sentences with different ASR-confidence. This technique is also complementary to linear discriminant analysis feature projection. An overall BLEU score improvement of 0.80 was achieved on an in-domain English-to-French SLT task.
Metadata
Item Type: | Proceedings Paper |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | © 2016 IEEE |
Keywords: | groupwise learning; ordinal regression; spoken language translation; support vector regression |
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Computer Science (Sheffield) |
Depositing User: | Symplectic Sheffield |
Date Deposited: | 10 Aug 2016 12:03 |
Last Modified: | 19 Dec 2022 13:34 |
Status: | Published |
Series Name: | http://dx.doi.org/10.1109/ICASSP.2016.7472853 |
Refereed: | Yes |
Identification Number: | 10.1109/ICASSP.2016.7472853 |
Related URLs: | |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:103466 |